Assessing the quality of Portuguese child care programs for toddlers

Sílvia Barros, Escola Superior de Educação, Instituto Politécnico do Porto; University of Porto, Department of

Psychology and Educational Sciences, Porto, Portugal

Cecília Aguiar, UIPCDE, Instituto Superior de Psicologia Aplicada, Lisboa, Portugal

Abstract

The purpose of this study was to describe the quality of toddler child care classrooms in the

district of Porto, in the north of Portugal. One hundred and sixty classrooms for children

between 1 and 3 years of age participated in this study. Results suggested the existence of

poor average quality and absence of good-quality classrooms. Child–adult ratio was a

statistically significant predictor of overall child care quality (with lower ratios associated with

higher quality). The results of this study provide information on the applicability of ITERS-R to a

new cultural and linguistic context, suggesting the need to enhance the quality of Portuguese

programs serving toddlers.

Keywords

Quality; Toddler child care; ITERS-R; Portugal

Portuguese families with children between 4 months and 3 years of age are served by four

types of child care and education programs: crèches (infant–toddler child care centers), mini-

crèches (infant–toddler child care for 12–15 children), official nannies (family child care homes),

and family crèches (groups of 12–20 family child care homes), all dependent on the Ministry of

Labor and Welfare (Vasconcelos, Orey, Homem, & Cabral, 2002). The number of child care and

education services for infants and toddlers has been progressively increasing in Portugal.

Coverage rate has increased from 12.65% in 1996 (Vasconcelos et al., 2002) to 21.5% in 2003

(Instituto da Segurança Social, 2005c).

Currently, project PARES (Programa de Alargamento da Rede de Equipamentos Sociais), a

governmental program designed to increase social protection and support families by extending

the network of social programs, is funding a 50% increase in the number of infant–toddler child

care centers ( Instituto da Segurança Social, 2005c). The goal is to achieve a 33% coverage

rate in 2010 ( Instituto da Segurança Social, 2005c). However, it is important that the

investment made in increasing the number of child care programs accessible to families with

young children is accompanied with an effort to assure their quality. This fact is acknowledged

by the Portuguese government agency responsible for infant and toddler child care in Portugal –

The Social Solidarity Institute – as suggested by the recent distribution of documents that

provide an Infant – Toddler Child Care Quality Assessment Model and a Key-Processes Manual

( Instituto da Segurança Social, 2005a and Instituto da Segurança Social, 2005b).

brought to you by COREView metadata, citation and similar papers at core.ac.uk

provided by Scientific Repository of the Polytechnic Institute of Porto

Several studies have consistently demonstrated an association between the quality of child care

programs and children's short-term and long-term developmental outcomes (National

Association for the Education of Young Children [NAEYC], 1997). For example, in the United

States of America (USA), the Cost, Quality, and Child Outcomes in Child Care Centers Study

(Peisner-Feinberg & Burchinal, 1997) found a positive association between children's cognitive

and social developmental outcomes and the quality of their experiences in child care programs,

after accounting for children's and families’ characteristics. Results further indicated that (a)

high-quality care positively predicts academic performance; (b) children who have traditionally

been considered as being “at risk” are more influenced by the quality of child care experiences;

(c) different aspects of quality influence different development areas — the quality of child care

classroom practices was related to children's cognitive development, while the closeness of the

teacher-child relationship influenced children's social development through the early school

years; (d) high-quality child care continues to positively predict children's performance over

time, in many cases through the end of second grade (Peisner-Feinberg and Burchinal,

1997 and Peisner-Feinberg et al., 2000). Another major USA study, the National Institute on

Child Health and Human Development [NICHD] Study of Early Child Care (NICHD, 2005), found

that lower quality child care predicted less harmonious mother–child relationships, increased

probability of insecure attachment of children with less sensitive mothers, lower language and

cognitive outcomes, and lower school readiness competencies.

These results highlight the need to assess and enhance the quality of child care and education

structures. In Portugal, few studies have examined the quality of care programs serving infants

and toddlers. In 2001, a small study (“Creches e jardins-de-infância”, 2001), based on the

inspection of 25 child care centers (20 of which serving infants and toddlers), found worrying

conditions in adult–child ratio, group size, outdoor and indoor facilities and equipments, and

hygiene and safety. Likewise, the OECD, 2000 and OECD, 2001expressed concern about the

quality of education and care programs serving Portuguese infants and toddlers, recommending

the implementation of self-assessment and inspection procedures. In 2002, Aguiar, Bairrão, and

Barros observed 30 toddler child care classrooms, in the district of Porto, reporting average

poor quality and absence of good-quality programs for toddlers. Using the same

participants,Pessanha (2005) found a small positive effect of toddler child care quality on

children's developmental outcomes.

Despite varying views on quality, there has been some consensus, at least in Europe, USA, and

other industrialized countries, around the thought that quality is related to the physical and

social characteristics of the settings where child care and education take place and to staff

characteristics (Bairrão, 1998). For example, NAEYC (1998) defined a high-quality program as

one that ensures a safe and stimulating environment that promotes children's physical, social,

emotional, aesthetic, intellectual, and language development, while being sensitive to families’

needs and preferences. According to Harms and Clifford (1993), the most agreed on criteria to

determine quality include health and safety (licensing criteria, compliance with sanitary and

health criteria); interactions (responsive and warm adult–child interactions); curriculum

(developmentally appropriate activities, balance of child-initiated/teacher-initiated activities,

etc.); and staff (limited group size, appropriate adult–child ratio, and staff training). Based on

these criteria,Harms, Cryer, and Clifford (1990) developed The Infant/Toddler Environment

Rating Scale – ITERS, a scale designed to assess the global quality of educational programs

serving children from birth to 30 months of age. The revised edition of ITERS (the Infant/Toddler

Environment Rating Scale – Revised Edition, ITERS-R) was published in 2003 (Harms, Cryer, &

Clifford, 2003) and translated into Portuguese in 2004 (Harms et al., 2003 and Harms et al.,

2004).

The definition of quality underlying ITERS and ITERS-R includes research's results,

professional values, and empirical knowledge, and is consistent with the Criteria for Quality

Early Childhood Programs, defined by the NAEYC (1998) and with the goals or requirements of

the Child Development Associate – CDA (CDA National Credentialing Program, 1984, as cited

in Harms, Cryer, & Clifford, 1999). The ITERS is considered a process measure (Vandell &

Wolfe, 2000), the quality dimension more directly experienced and felt by children (Cryer &

Burchinal, 1997). According to Vandell and Wolfe (2000), child care processes relate to the

experiences that occur in child care settings, including children's interactions with caregivers

and peers and their participation in different activities.

Research has demonstrated that the ITERS/ITERS-R and/or the Early Childhood Environment

Rating Scale – ECERS (Harms & Clifford, 1980) and its revised version (ECERS-R, Harms,

Clifford, & Cryer, 1998) correlate with other indicators of child care quality, namely structural

characteristics (e.g., Cost et al., 1995 and Vandell and Wolfe, 2000) such as staff–child ratio,

numbers of adults and children in the classroom, caregiver education, training, age, wage, and

years of teaching experience, and type of center (i.e., for-profit or non-profit). Higher quality has

generally been observed in classrooms with less children per adult (e.g., Cost et al.,

1995, Cryer et al., 1999 and Phillips et al., 2000), smaller group sizes (e.g.,ECCES,

1997, Gevers Deynoot-Schaub and Riksen-Walraven, 2005 and Goelman et al., 2000), and

more adults (Goelman et al., 2000). However, some studies did not confirm the association

between process quality and some of these specific structural dimensions. For instance, Gevers

Deynoot-Schaub and Riksen-Walraven (2005) did not find an association between ITERS

overall score and adult–child ratio and, in Portugal, Abreu-Lima and Nunes (2006) did not find

an association between ECERS-R scores and group size or adult–child ratio. Children's age

group has not been found to correlate with preschool quality global scores in Portugal,

Germany, and Spain (European Child Care and Education Study Group, 1997). However, in

toddler child care classrooms, higher quality has been observed in classrooms for older toddlers

(Aguiar, Bairrão, & Barros, 2002). The association between process quality measures and

teachers’ age and experience is not consistent, with studies showing no associations

(e.g., Gevers Deynoot-Schaub & Riksen-Walraven, 2005), negative associations (Pessanha,

2005 and van Ijzendoorn et al., 1998), and positive associations (e.g., Bryant, Burchinal, Lau, &

Sparling, 1994). On the other hand, research suggests teachers’ wages may be one of the

strongest correlates of process quality (e.g., Phillips et al., 2000). Another clear positive

association has been found between global quality and teachers’ education and training, with

more highly educated and trained teachers providing higher quality environments than less

educated and trained teachers (e.g., Aguiar et al., 2002, Cost et al., 1995 and Goelman et al.,

2000). Although less frequently studied, the type of center was found to correlate with global

quality, with non-profit or subsidized centers scoring higher than for-profit centers on process

quality measures (e.g., Aguiar et al., 2002 and Gevers Deynoot-Schaub and Riksen-Walraven,

2005).

In a national context characterized by (1) a great demand for structures that allow families to

conciliate work and family life, (2) a scarce supply of education and care services for infants and

toddlers, and (3) empirical indicators of insufficient quality of such services, there is a real

necessity to thoroughly examine the characteristics of programs serving children in early stages

of development. Therefore, the main goals of this study were (1) to investigate the applicability

of ITERS-R to a new culture and language environment by examining the psychometric

properties of data collected using the Portuguese translation, (2) to describe the characteristics

of child care and education settings for children under 3 years of age, and (3) to examine the

associations among overall process quality and selected structural features of toddler child care

classrooms. While the first two goals are clearly descriptive and, thus, no research hypotheses

were formulated, the following hypotheses specify our expectations regarding the third goal:

process quality would be higher (a) in non-profit than in for-profit centers, (b) in classrooms with

lower child–adult ratios than in classrooms with higher child–adult ratios, (c) in classrooms with

a lead teacher with a university degree in early childhood education than in classrooms with a

lead adult without a university degree, (d) when teachers receive higher wages, and (e) in

classrooms serving older toddlers than in classrooms serving younger toddlers.

1. Method

1.1. Participants

One hundred and sixty toddler child care classrooms, from 160 centers randomly selected from

a list of child care centers in the district of Porto, participated in this study. Porto is the second

largest city in Portugal, with a metropolitan area of approximately 1,500,000 inhabitants

(Instituto Nacional de Estatística, 2005). We used stratified random selection in order to select

80 non-profit private centers and 80 for-profit private centers. The participation rate for non-

profit centers was 96% and the participation for for-profit centers was 69%.

In each child care center, we observed a single classroom, in a total of 65 classrooms for

children between 1 and 2 years old, 80 classrooms for children between 2 and 3 years old, and

15 classrooms for children between 1 and 3 years old. Although classrooms serving infants

(from the age of 4 months) were available, they were not included in the present study because

collection of additional measures not reported here was not possible in this age group.

The number of children in each classroom ranged from 4 to 30 (M = 12.94, SD = 4.62) and the

number of adults varied between 1 and 5 (M = 2.02, SD = 0.76). Child–adult ratio ranged from

2:1 to 17:1 (M = 6.97:1,SD = 2.82:1). These classrooms included 323 adults and served 2070

children.

In 132 classrooms, the lead adult (i.e., the adult responsible for providing all or most of the

direct work with children) was a trained teacher (i.e., a teacher with a college degree in early

childhood education), and in 28 classrooms the lead adult was an untrained teacher (i.e., a

teacher's aid without a college degree). All lead adults will be named teachers during the course

of this paper.

Teachers’ education ranged from 5 to 18 years of formal schooling (M = 14.88, SD = 2.41), with

untrained teachers reporting between 5 and 16 years of education and trained teachers

reporting having a bachelors’ degree (n = 32), a licentiate degree (n = 97), or a post-graduate

degree (n = 3). Teachers’ age ranged from 21 to 52 years of age (M = 32.27, SD = 6.93) and

teachers’ experience ranged from 2 months to 32 years (M = 8.09, SD = 6.66).

1.2. Instruments

Global quality of individual classrooms was assessed using the Portuguese translation of

theInfant/Toddler Environment Rating Scale–Revised Edition ( Harms et al., 2003). The

translation of ITERS-R used in this study was conducted by the first author, under the

supervision of a senior scientist specialized in the field of early childhood education, and is

currently under revision for publication purposes. The ITERS-R consists of 39 items, organized

under seven categories or conceptually defined subscales: I – Space and Furnishings, II –

Personal Care Routines, III – Listening and Talking, IV – Activities, V – Interaction, VI – Program

Structure, and VII – Parents and Staff. Each item is presented as a 7-point scale, with

descriptors for 1 (inadequate), 3 (minimal), 5 (good), and 7 (excellent).

Using the procedure adopted by other studies (e.g., Tietze, Cryer, Bairrão, Palacios, & Wetzel,

1996), due to the presence of missing values, permitted by the scale, items 21 (sand and water

play), 23 (use of TV, video, and/or computer), 31 (group play activities), 32 (provisions for

children with disabilities), and 36 (staff interaction and cooperation) were excluded from

analyses. Cronbach's alpha coefficient on the overall scale was .75 when using all 39 items and

.83 when using 34 items. These results suggest that a single construct is measured by ITERS-

R. Similar (see Vermeer et al., 2008) and higher (see Bisceglia, Perlman, Schaack, & Jenkins,

2009) alpha coefficients were previously reported for the ITERS-R overall score.

We also computed internal consistency coefficients for data obtained in each of the

conceptually defined subscales (see Table 1). Alpha coefficients are lower than those reported

by Bisceglia et al. (2009) for all subscales and suggest caution should be used when conducting

analyses at the subscale level.

Information on child care structural features was collected using a questionnaire designed for

that purpose. We collected data on group size, number of adults in the classroom, type of

center (for-profit or non-profit), children's age group, and lead teacher's education.

1.3. Procedure

1.3.1. Training

Previous to the ITERS-R translation and use, the first author participated in an intensive

training, lead by the authors of the scale, at the University of North Carolina at Chapel Hill. Later

on, two other observers were trained by the first researcher using the materials recommended

by the authors of the scale, namely the training video (Harms & Cryer, 2003). Following this

procedure, the three observers conducted training sessions in 16 toddler child care classrooms,

aiming at 85% interrater agreement, within one scale point. The mean interrater percent

agreement, within one scale point, for these 16 classrooms, was 88%.

1.3.2. Data collection

All data were collected between September 2004 and October 2005, by these three

researchers. As suggested by scale authors, each observer remained with the group of children

for at least 3 h in order to observe classroom main routines and activities, between arrival and

nap. All classrooms were observed during the morning, the period of greatest activity among

toddler child care classrooms. Following the observation of each classroom, observers

conducted a small interview with the lead teacher, in order to score all indicators in each item

and to collect information on classroom structural features.

Interobserver agreement checks were conducted during the data collection procedure, across

44 agreement sessions (27.5% of the 160 classrooms). Interrater exact percent agreement

ranged from 43.18% to 95.45% (M = 77.32, SD = 14.82) and interrater within one scale point

percent agreement ranged from 84.09% to 100% (M = 95.87, SD = 4.55). Weighted kappa

ranged from −.03 to .95 (M = .69, SD = .24).

2. Results

Overall mean results on ITERS-R ranged from 1.62 to 4.09 (M = 2.84, SD = .48), presenting a

normal distribution. Sixty one percent of classrooms were given scores that suggest inadequate

quality (i.e., with an overall score under 3) and only 39% of classrooms were given scores that

indicated minimal quality (i.e., with an overall score between 3 and 5). Within the participating

classrooms, there were no high-quality classrooms or, in other words, classrooms that

consistently implemented developmentally appropriate practices (i.e., with an overall score

equal to or greater than 5).

Table 2 presents descriptive statistics for ITERS-R subscales and items. Mean subscale scores

varied between 1.67 (Personal Care Routines) and 3.70 (Interaction). Three of the seven

subscales presented mean scores below 3 (i.e., minimal quality) and none of the subscales

scored higher than 4.

Mean results at the item level ranged from 1.06 (diapering/toileting) to 4.41 (staff continuity).

About half of the items presented mean scores that indicate the presence of minimal quality,

while the remaining half presented mean scores that suggest the provision of poor quality child

care. Twenty-one of the 34 items were scored between 1 and 7. Of the remaining 13 items, all

presented minimal ratings of 1, except for item number 5 (display for children). Regarding

maximum scores, items 9 (diapering/toileting), 10 (health practices), 11 (safety practices), 22

(nature/science), and 24 (promoting acceptance of diversity) scored 4, item 7 (meals/snacks)

scored 5, and items 5 (display for children), 12 (helping children understand language), 16

(active physical play), 18 (music and movement), 19 (blocks), 20 (dramatic play), and 26 (peer

interaction) scored 6. Maximum scores obtained in some items may raise some concern

regarding their sensitivity.

Kurtosis and skewness values for items 8, 9, 14, 16, 24, 29, 30, 34, and 39 indicate the violation

of the normality assumption. Once more, and excluding results obtained on item 34 (an item

with greater score variability and higher mean scores), these results seem to reflect

concentration of scores on lower scale points due to lack of good-quality care in observed

classrooms.

In order to examine the associations among overall process quality and structural features of

toddler child care classrooms (i.e., teachers’ professional status/training, children's age group—

determined on the base of the centers’ distribution of children across same-age classrooms and

excluding the 15 classrooms for children between 1 and 3 years of age, type of center, child–

adult ratio, and teachers’ monthly wage) we used standard multiple regression analyses. All

variables were entered in the model in one step as our hypotheses did not specify subtypes of

structural features or the relative contribution of structural variables measured at different levels.

Variables included in the regression model were either dichotomous (teachers’ professional

status, children's age group, and type of center), or normally distributed interval variables.

Teachers’ professional status was coded 0 for trained teachers and 1 for untrained teachers,

type of center was coded 0 for non-profit centers and 1 for for-profit centers, and age group was

coded 0 for 1-year-old classrooms and 1 for 2-year-old classrooms. Missing values on teacher

monthly income (validN = 110) and elimination of two outliers for the child–adult ratio variable

limited the number of classrooms included in the regression model to 108. A logarithm

transformation was performed on child–adult ratio to improve normality. For descriptive

purposes, correlation coefficients between classroom overall quality and structural predictors,

and among predictors, are presented in Table 3. Regression results (see Table 4) suggest only

child–adult ratio was a statistically significant predictor of overall process quality, after

accounting for the effects of the remaining variables [R = .40, F(5, 102) = 3.85, p < .01]. Effect

size estimates suggest a medium effect of the model (f2 = .19).

We also computed Spearman's rho correlations between classroom quality and classroom

structural features that were not normally distributed (and normality could not be achieved

through transformations), such as group size, number of adults in the classroom, teachers’

education (years), teachers’ age, and teachers’ experience. We found small statistically

significant positive associations between overall quality and group size (rs = .23, p = .003) and

teachers’ education (rs = .29, p = .000); and a moderate positive association between quality

and number of adults in the classroom (rs = .41, p = .000). Note that both group size and

number of adults in the classroom were moderately associated with type of center, with bigger

groups attending non-profit centers (r = −.44, p = .000) and less adults working in for-profit

centers (r = −.56, p = .000).

3. Discussion

Results from this study suggest that toddlers attending child care classrooms in the district of

Porto, Portugal, are receiving poor quality care and education that does not appropriately

respond to children's basic health, safety, and developmental needs. Only 39% of the observed

classrooms provided sufficient quality that minimally met custodial care needs and, to some

small degree, responded to children's basic developmental needs. Results further suggest the

absence of developmentally appropriate practices or good quality in programs serving children

from this age group.

These findings support results previously reported by Aguiar et al. (2002) and strengthen the

empirically based concerns about the characteristics of extra-familiar settings available to

families with young children in Portugal. Although particularly salient in the Portuguese context

(given the apparent absence of high-quality toddler child care), concerns about the quality of

child care and education programs for children under 3 years of age are felt in other countries.

For example, in the USA, according to the Cost et al. (1995), only 8% of the classrooms for

children under 3 years of age presented high quality, capable of promoting children's health and

cognitive and social behavior, opposed to 40% of classrooms that presented low or

inappropriate quality, considered eventually dangerous for children's health and safety, and

damaging to children's cognitive and social development. More recent data on infant and toddler

child care quality in the USA indicated that only 9% of children attended excellent quality

programs and 29.6% attended good-quality programs, whereas 53.2% of children attended

mediocre quality programs and 8.1% attended low-quality programs (NICHD, 2000 and NICHD,

2005). Nevertheless, mean scores obtained in the present study (M = 2.84) are substantially

lower than those recently reported by Hestenes, Cassidy, Hegde, and Lower (2007). Using the

same quality measure and a larger sample, these authors reported a global quality score that

suggests average quality care and education is available to American infants and toddlers

(M = 4.89). Within the European context, and considering studies that used ITERS or ITERS-R,

we highlight the Dutch results reported by Vermeer et al. (2008), showing that although in 1995

no center received low level quality scores, in 2001, 6% of centers were reported to provide low-

quality child care services, while, in 2005, the percentage of low-quality centers increased to

36%. On the other hand, no group received good or excellent quality ratings in 2001, similarly to

the Portuguese sample. In Greece, results were not so different: none of the child care centers

assessed using the ITERS were found to have high-quality scores, but rather ranged between “

minimal” and “good” quality ( Petrogiannis, 2002). Better results were found in the United

Kingdom (UK), where 23% of centers achieved a total score of 5 or higher, with only 7% scoring

less than 3 ( Mathers & Sylva, 2007). Mathers and Sylva believe the relatively high-quality

ratings may be due to the legislative regulation in the UK and also to the additional resources

participating centers had access to, due to involvement in the Neighbourhood Nursery Initiative.

Despite the similarities between Portugal and some other European countries regarding the

nonexistence of centers with good or excellent scores, the Portuguese mean scores are lower

than those from the Netherlands (M = 3.40) ( Vermeer et al., 2008), Greece (M = 3.50)

( Petrogiannis, 2002), Germany (M = 3.21) ( Tietze & Cryer, 2004), and UK (M = 4.40) ( Mathers

& Sylva, 2007); the magnitude of differences suggests real-life practical effects, with Portuguese

children experiencing lower quality experiences, when compared to children from other

European countries and USA.

Analyses of results at the item level allow us to elaborate on a profile of toddler child care

classrooms in the district of Porto. Areas of major concern, that is, dimensions where quality of

care is most inappropriate, included (a) health requirements, (b) hygiene procedures before and

during meals, diapering, and toileting, (c) availability and use of books, (d) frequency of active

outdoor physical play, (e) availability of blocks, (f) availability of nature and science materials

and activities, and (g) encouragement of acceptance of cultural, ethnic, ability, and role

diversity. Low scores on such items may, among other disadvantages, place children at

increased risk of infections and limit children's opportunities to develop language and literacy

skills, critical scientific reasoning, and positive attitudes toward diversity. In contrast, relative

strengths of participating classrooms, that is, dimensions where quality of care ensured minimal

response to children's developmental needs, included (a) materials and activities to promote

receptive and expressive language, (b) availability of fine motor materials and activities, (c)

promotion of peer interaction, (d) continuity of staff, and (e) provisions for the professional

development of staff members.

Results on the item level are reinforced by subscale results, although we should be cautious at

this level given the low internal consistency values. In fact, Personal care routines, Activities,

and Program structure had the lowest results, and did not reach the minimal quality

rating. Vermeer et al. (2008) also reported the lowest scores for the subscales of Personal care

routines and Activities and higher scores for the Interaction subscale in the Netherlands. Similar

results were found in the UK but not in Greece, where the ITERS subscale Personal care

routines had the second highest mean (M = 4.0), following the Interaction subscale

( Petrogiannis, 2002). These results may suggest ITERS-R is more demanding at the health,

hygiene, materials, and activities requirements or, for some reason, centers and teachers are

less prepared to assure those requirements. Mathers, Linskey, Seddon, and Sylva

(2007) concluded that the ECERS and ITERS approach to personal care and hygiene is often

more demanding than the UK National Standards. In Portugal, although parents and teachers

agree on the high importance of those items (Barros, 2007), the national legislation and

orientations are too broad to provide specific guidance.

The small to moderate statistically significant associations among child care quality and

selected structural features of the participating classrooms seem to mirror the relative

importance of physical and organizational characteristics of child care programs when

considering the global quality of children's educational experiences. Interestingly, associations

among child care quality and group size, number of adults, and child–adult ratio suggest that

classrooms with a larger number of children and a larger number of adults are given higher

quality scores, although classrooms with smaller child–adult ratios are given higher quality

scores. Regression analyses suggested that child–adult ratio was in fact a statistically

significant predictor of overall quality, after accounting for the effects of other structural

variables. Due to data distribution properties, we could not include important structural features

in the regression analyses. However, Spearman's correlation coefficients suggest that the

unexpected positive association between group size and overall quality may be due to type of

center, considering that in this study, non-profit centers had bigger groups of children but also

had more teachers with a university degree, lower child-staff ratios, and higher wages than for-

profit centers.

ITERS-R has been used in several countries and with different purposes, namely in research

and quality enhancement programs (Harms et al., 2003). Results obtained in the present study

may contribute to substantiate the use of such a quality measure in the assessment of

Portuguese toddler child care classrooms. Despite the low mean results on ITERS-R, we found

a normal distribution, with appropriate skewness and kurtosis. Although mean scores at the item

level concentrate at the lower end of the likert scale used in ITERS-R and these results could be

attributed to problems in item design and/or to strict requirements in the scoring system, a

qualitative analysis of the observation records suggests that such findings mirror the actual low-

quality of child care provision. Thus, it is important to emphasize that (1) hygiene problems

during diapering and toileting as well as during meals and snacks (particularly regarding adult

and child hand washing) were frequently observed; (2) in a considerable number of classrooms

books were not accessible for children; (3) outdoor space was not frequently used (a

considerable number of teachers only provided outdoor activities in the spring); and (4) centers

did not promote in-service training for staff members.

In conclusion, the present study expands the evidence base on toddler child care quality,

providing information on the applicability of ITERS-R to a new culture and language

environment, and simultaneously suggesting the need to make a deliberate investment in

quality assessment and enhancement. It is outside of the scope of this paper and beyond the

inferences warranted by our data to provide definite explanations as to why Portuguese child

care classrooms serving toddlers present lower mean scores on the ITERS/ITERS-R than the

Netherlands, Germany, UK, Greece, and the USA. Such differences in the quality of child care

and education services are, most likely, multidetermined and Portuguese early childhood

education researchers are well advised to analyze the foundations for the present level of

quality in toddler child care classrooms. Tentatively, we argue that these differences may be

partially related to (1) the historically, legally, and politically determined “caretaking” nature of

the services provided to groups of infants and toddlers, estranged from an educational

perspective; (2) the impoverishing legacy of a society that 35 years ago had high illiteracy rates

and that, although increasingly educated, still has education rates below average in OECD

(OECD, 2006); (3) the low (although increasing) critical mass of scientific production in the field

of early childhood development and education; (4) the nonexistence of a coordinated vision for

early childhood that brings together education, welfare, health, employment, and family policies

(Alarcão, 2008); and (5) the insufficient resources of a society that struggles with poverty rates

above the mean of OECD countries (Instituto Nacional de Estatística, 2006).

As suggested by our findings, teacher training and education seem to constitute necessary

pathways to improve toddler child care quality. A considerable proportion of lead teachers in

toddler classrooms do not have a college degree in Early Childhood Education and the lead

teachers that do have a college degree in early child care and education received training that

was clearly focused on preschool-aged children (that is, on working with children between 3 and

5), neglecting child care practices focusing on infants and toddlers.

The improvement of the quality of care provided to younger children may be helped if parents

can identify good-quality child care centers (Cost et al., 1995). Available data suggest that

Portuguese families that chose infant–toddler child care as an education and care resource are

limited in their selection capacity for a number of reasons: (1) the limited number of programs

available (Instituto da Segurança Social, 2005c), (2) the relative homogeneous quality of

existing classrooms (Aguiar et al., 2002), (3) the lack of relevant information on quality criteria

for infant–toddler child care, and (4) the impossibility to access results of quality assessments by

independent observers (Cryer, Tietze, & Wessels, 2002). Considering this picture, necessary

direct intervention strategies aiming at improving child care settings’ structural and process

characteristics must be complemented with systematic efforts aiming at helping parents become

more discriminating in their choices, and therefore, increasing their power to affect child care

settings quality (Cryer et al., 2002). Furthermore, centers should be actively engaged in the

effort to enhance the quality of programs serving toddlers and, as stated by Mathers et al.

(2007), ITERS-R should be used “with” centers to encourage the self-evaluation process and

not simply applied “to them” as a monitoring tool.

References

Abreu-Lima and Nunes, 2006 Abreu-Lima, I., & Nunes, C. (2006). A Escala de Avaliação do

Ambiente em Educação de Infância–Versão Revista [The Early Childhood Environment Rating

Scale–Revised Edition]. In: C. Machado, L. Almeida, M.A. Guisande, M. Gonçalves, V. Ramalho

(Coord.),Avaliação Psicológica: Formas e Contextos. Vol. XI. (pp. 634–643). Braga:

Psiquilíbrios Edições.

Aguiar et al., 2002 C. Aguiar, J. Bairrão, S. Barros Contributos para o estudo da qualidade em

contexto de creche na Área Metropolitana do Porto [Contributions for the study of toddler child

care quality in the metropolitan area of Porto] Infância e Educação: Investigação e Práticas, 5 (2002),

pp. 7–28

Alarcão, 2008 I. Alarcão Considerações finais e recomendações do estudo [Final

considerations and study recommendations] Conselho Nacional de Educação (Ed.), A educação das

crianças dos 0-12 anos, Conselho Nacional de Educação (2008), pp. 114–127 Retrieved

from http://www.cnedu.pt/files/ESTUDO.pdf

Bairrão, 1998 J. Bairrão O que é a qualidade em educação pré-escolar? [What is quality in

preschool education?] Ministério da Educação (Ed.), Qualidade e projecto na educação pré-escolar,

Ministério da Educação (1998), pp. 43–88

Barros, 2007 Barros, S. (2007). Qualidade em contexto de creche: Ideias e práticas [Quality in

infant–toddler child care: Ideas and practices]. Unpublished doctoral dissertation, University of

Porto.

Bisceglia et al., 2009 R. Bisceglia, M. Perlman, D. Schaack, J. Jenkins Examining the

psychometric properties of the Infant–Toddler Environment Rating Scale-Revised Edition in a

high-stakes context Early Childhood Research Quarterly, 24 (2009), pp. 121–132

Bryant et al., 1994 D.M. Bryant, M. Burchinal, L.B. Lau, J.J. Sparling Family and classroom

correlates of Head Start children's developmental outcomes Early Childhood Research Quarterly, 9

(1994), pp. 289–309

Cost et al., 1995 Cost, Quality, Child Outcomes Study Team Cost, quality, and child outcomes

in child care centers: Public report (2nd edition)Economics Department, University of Colorado at

Denver, Denver (1995)

Creches, 2001 Creches e jardins-de-infância [Infant–toddler and preschool child care]. (2001,

July/August). Proteste, 216, 6–11.

Cryer and Burchinal, 1997 D. Cryer, M. Burchinal Parents as child care consumers Early

Childhood Research Quarterly, 12 (1997), pp. 35–38

Cryer et al., 1999 D. Cryer, W. Tietze, M. Burchinal, T. Leal, J. Palacios Predicting process

quality from structural quality in preschool programs: A cross-country comparison Early Childhood

Research Quarterly, 14 (3) (1999), pp. 339–361

Cryer et al., 2002 D. Cryer, W. Tietze, H. Wessels Parents’ perceptions of their children's child

care: A cross-national comparison Early Childhood Research Quarterly, 17 (2002), pp. 259–277

ECCES, 1997 European Child Care and Education Study Group. (1997). European Child Care

and Education study: Cross national analyses of the quality and effects of early childhood

programmes on children's development. Final report for work package #1 submitted to

European Union DG XII: Science, Research and Development RTD Action: Targeted Socio-

Economic Research. Unpublished Report.

Gevers Deynoot-Schaub and Riksen-Walraven, 2005 M.J.J.M. Gevers Deynoot-Schaub, J.M.A.

Riksen-Walraven Child care under pressure: The quality of Dutch centers in 1995 and in 2001

The Journal of Genetic Psychology, 166 (2005), pp. 280–296

Goelman et al., 2000 Goelman, H., Doherty, G., Lero, D. S., LaGrange, A., & Tougas, J.

(2000). You bet I care. Care and learning environments: Quality in child care centres across

Canada. Retrieved 25 June 2004, from http://uoguelph.ca/cfww.

Harms and Clifford, 1980 T. Harms, R.M. Clifford The Early Childhood Environment Rating

Scale Teachers College Press, New York (1980)

Harms and Clifford, 1993 T. Harms, R.M. Clifford Studying educational settings Spodek (Ed.),

Handbook of research on the education of young children, Macmillan Publishing Company, New York

(1993), pp. 477–491

Harms et al., 1998 T. Harms, R.M. Clifford, D. Cryer The Early Childhood Environment Rating

Scale (Revised Edition)Teachers College Press, New York (1998)

Harms and Cryer, 2003 T. Harms, D. Cryer Video observations for the Infant/Toddler

Environment Rating Scale (Revised Edition)Teachers College Press, New York (2003)

Harms et al., 1990 Harms, T., Cryer, D., & Clifford, R. M. (1990). Infant/Toddler Environment

Rating Scale (ITERS). Frank Porter Graham Child Development Center. University of North

Carolina at Chapel Hill.

Harms et al., 1999 Harms, T., Cryer, D., & Clifford, R. M. (1999). Infant/Toddler Environment

Rating Scale. Retrieved July 2002, from www.fpg.unc.edu/∼ecers/iters.html.

Harms et al., 2003 T. Harms, D. Cryer, R.M. Clifford Infant/Toddler Environment Rating Scale –

Revised Edition (ITERS-R) Teachers College Press, New York (2003)

Harms et al., 2004 Harms, T., Cryer, D., & Clifford, R. M. (2004). Infant/Toddler Environment

Rating Scale–Revised Edition (ITERS-R) (S. Barros, Trans.). Unpublished translation (Original

work published 2003).

Hestenes et al., 2007 L.L. Hestenes, D.J. Cassidy, A.V. Hegde, J.K. Lower Quality in inclusive

and noninclusive infant and toddler classrooms Journal of Research in Childhood Education, 22 (1)

(2007), pp. 69–84

Instituto da Segurança Social, 2005a Instituto da Segurança Social. (2005a). Manual de

processos-chave: Creche [Key-processes manual: Infant–toddler child care]. Ministério da

Segurança Social e do Trabalho.

Instituto da Segurança Social, 2005b Instituto da Segurança Social. (2005b). Modelo de

avaliação da qualidade: Creche [Infant–toddler child care quality assessment model]. Ministério

da Segurança Social e do Trabalho.

Instituto da Segurança Social, 2005c Instituto da Segurança Social. (2005c). Relatório de

actualização 2005–2006 do Plano Nacional para a Inclusão [2005–2006 update report of the

National Inclusion Plan]. Retrieved 27 June, 2006,

from http://www.dgeep.mtss.gov.pt/estudos/peis/rapnai0506_pt.pdf.

Instituto Nacional de Estatística, 2005 Instituto Nacional de Estatística. (2005). Grande Área

Metropolitana do Porto [Great Metropolitan Area of Porto]. Retrieved December 2006,

from http://www.ine.pt/prodserv/Edi_Bolso/pdf/2005/Norte/GaMP_web.pdf.

Instituto Nacional de Estatística, 2006 Instituto Nacional de Estatística. (2006,

December). Indicadores sociais, Edição de 2005. Destaque. Informação à Comunicação

Social [Social indicators, 2005 Edition. Highlights. Information to the media]. Retrieved 31

January, 2007,

from http://www.ine.pt/portal/page/portal/PORTAL_INE/Destaques?DESTAQUESdest_boui=746

39&DESTAQUESmodo=2.

Mathers et al., 2007 S. Mathers, F. Linskey, J. Seddon, K. Sylva Using quality rating scales for

professional development: Experiences from the UK International Journal of Early Years Education,

15 (3) (2007), pp. 261–274

Mathers and Sylva, 2007 Mathers, S., & Sylva, K. (2007). National evaluation of the

Neighbourhood Nurseries Initiative: The relationship between quality and children's behavioural

development. SureStart Research Report SSU/2007/FR/022. Retrieved 21 May, 2009

from http://www.dcsf.gov.uk/research/data/uploadfiles/SSU2007FR022%20REV.pdf.

NAEYC, 1997 National Association for the Education of Young Children Developmentally

appropriate practice in early childhood programs serving children from birth trough age 8

NAEYC, Washington (1997)

NAEYC, 1998 National Association for the Education of Young Children Accreditation criteria

and procedures: Summary of a position statement of the National Association for the Education

of Young Children NAEYC, Washington (1998)

NICHD, 2000 National Institute on Child Health and Human Development. (2000). The NICHD

Study of Early Child Care. Retrieved 14 June, 2002

from http://www.childresearch.net/CYBRARY/EDATA/NICHD/DATA03.HTM.

NICHD, 2005 National Institute on Child Health and Human Development Early Child Care

Research Network Child care and child development: Results from the NICHD Study of Early

Child Care and Youth Development The Guilford Press, New York (2005)

OECD, 2000 Organization for Economic Co-operation and Development OECD country note:

Early childhood education and care policy in Portugal Organization for Economic Co-operation and

Development, Paris (2000) Retrieved from http://www.oecd.org/dataoecd/52/30/2534928.pdf

OECD, 2001 Organization for Economic Co-operation and Development Starting strong: Early

childhood education and care Organization for Economic Co-operation and Development, Paris (2001)

OECD, 2006 Organization for Economic Co-operation and Development (2006). Economic

survey of Portugal, 2006. Policy Brief. Retrieved 5 May, 2006

from http://www.oecd.org/document/63/0,2340,en_2649_34569_36482175_1_1_1_1,00.html.

Peisner-Feinberg and Burchinal, 1997 E.S. Peisner-Feinberg, M.R. Burchinal Relations

between preschool children's child-care experiences and concurrent development: The Cost,

Quality, and Outcomes Study Merrill-Palmer Quarterly, 43 (1997), pp. 451–477

Peisner-Feinberg et al., 2000 E.S. Peisner-Feinberg, M.R. Burchinal, R.M. Clifford, M.L. Culkin,

C. Howes, S.L. Kagan, et al. The children of the Cost, Quality, and Outcomes Study go to

school: Technical Report University of North Carolina at Chapel Hill, Frank Porter Graham Child

Development Center, Chapel Hill (2000)

Pessanha, 2005 Pessanha, M. (2005). Vulnerabilidade e resiliência no desenvolvimento dos

indivíduos: Influência da qualidade dos contextos de socialização no desenvolvimento das

crianças[Vulnerability and resilience in the development of individuals: Influence of socialization

contexts in children's development]. Unpublished doctoral dissertation, University of Porto.

Petrogiannis, 2002 K. Petrogiannis Greek day care centres’ quality, caregivers’ behaviour and

children's development International Journal of Early Years Education, 10 (2002), pp. 137–148

Phillips et al., 2000 D. Phillips, D. Mekos, S. Scarr, K. McCartney, M. Abbott-Shim Within and

beyond the classroom door: Assessing quality in child care centers Early Childhood Research

Quarterly, 15 (4) (2000), pp. 475–496

Tietze and Cryer, 2004 W. Tietze, D. Cryer Comparisons of observed process quality in

German and American infant–toddler programs International Journal of Early Years Education, 12

(2004), pp. 43–62

Tietze et al., 1996 W. Tietze, D. Cryer, J. Bairrão, J. Palacios, G. Wetzel Comparisons of

observed process quality in early child care and education programs in five countries Early

Childhood Research Quarterly, 11 (1996), pp. 447–475

van Ijzendoorn et al., 1998 M.H. van Ijzendoorn, L.W.C. Taveccio, G.J.J.M. Stams, M.J.E.

Verhoeven, E.J. Reiling Quality of center day care and attunement between parents and

caregivers: Center day care in cross-national perspective The Journal of Genetic Psychology, 159

(4) (1998), pp. 437–454

Vandell and Wolfe, 2000 D.L. Vandell, B. Wolfe Child care quality: Does it matter and does it

need to be improved? Institute for Research on Poverty. University of Wisconsin-Madison, Washington

(2000) Retrieved 20 June, 2002 from http://aspe.os.dhhs.gov/hsp/ccquality00/ccqual.htm

Vasconcelos et al., 2002 Vasconcelos, T., Orey, I., Homem, L. F., & Cabral, M.

(2002). Educação de infância em Portugal: Situação e contextos numa perspectiva de

promoção de equidade e combate à exclusão [Early childhood education in Portugal: Situation

and contexts in a perspective of promotion of equity and eradication of exclusion]. Conselho

Nacional de Educação.

Vermeer et al., 2008 H.J. Vermeer, M.H. van IJzendoorn, R.E.L. de Kruif, R.G. Fukkink, L.W.C.

Tavecchio, J.M. Riksen-Walraven, et al. Child care in the Netherlands: Trends in quality over

the years 1995–2005 The Journal of Genetic Psychology, 169 (2008), pp. 360–385